Prediction of cardiopulmonary events using the STOP-Bang questionnaire in patients undergoing bronchoscopy with moderate sedation

Dynamic prediction of hypoxemia risk at different time points based on preoperative and intraoperative features: machine learning applications in outpatients undergoing esophagogastroduodenoscopy

BACKGROUND

Hypoxemia often occurs in outpatients undergoing anesthesia-assisted esophagogastroduodenoscopy (EGD). However, there is a scarcity in tools to predict the hypoxemia risk. We aimed to solve this problem by developing and validating machine learning (ML) models based on preoperative and intraoperative features.

METHODS

All data were retrospectively collected from June 2021 to February 2022. The most appropriate predictive features were selected by the least absolute shrinkage and selection operator, which were incorporated and modelled by 4 ML algorithms. The area under the precision-recall curve (AUPRC) was used as the main evaluation metric to select the best models, and the selected models were compared with the STOP-BANG score. Their predictive performance was visually interpreted by SHapley Additive exPlanations. The primary endpoint of this study was hypoxemia during the procedure, defined as at least one reading of pulse oximetry < 90% without probes misplacement from the anesthesia induction beginning to the end of EGD, while the secondary endpoint was hypoxemia during induction, from the induction beginning to the start of endoscopic intubation.

RESULTS

Of 1160 patients in the derivation cohort, 112 patients (9.6%) developed intraoperative hypoxemia, of which 102 (8.8%) occurred during the induction period. In temporal and external validation, no matter whether based on preoperative variables or still based on preoperative plus intraoperative variables, our models showed excellent predictive performance for the two endpoints, significantly better than STOP-BANG score. In the model interpretation section, preoperative variables (airway assessment indicators, pulse oximeter oxygen saturation and BMI) and intraoperative variables (the induced propofol dose) made the highest contribution to the predictions.

To our knowledge, our ML models were the first to predict hypoxemia risk, which achieved excellent overall predictive ability integrating various clinical indicators. These models have the potential to become an effective tool for adjusting sedation strategies flexibly and reducing the workload of anesthesiologists.KEY MESSAGESThis study is the first model employing ML methods based on preoperative and preoperative plus intraoperative variables for predicting the risk of hypoxemia during induction and the whole EGD procedure respectively.Our four models achieved satisfactory predictive performance and outperformed STOP-BANG score in terms of AUPRC in the temporal and external validation cohorts respectively.We found that the relevant variables of airway assessment should be fully taken into account when analyzing the risk factor of hypoxemia, and the effect of patients' age on their hypoxemia risk should be considered in conjunction with the propofol dose.

An interpretable artificial neural network model for predicting hypoxemia via an online tool in adult (18-64) patients during esophagogastroduodenoscopy

Background

The hypoxemia risk in adult (18-64) patients treated with esophagogastroduodenoscopy (EGD) under sedation often poses a dilemma for anesthesiologists. We aimed to establish an artificial neural network (ANN) model to solve this problem, and introduce the Shapley additive explanations (SHAP) algorithm to further improve the interpretability.

Methods

The relevant data of patients underwent routine anesthesia-assisted EGD were collected. Elastic network was used to filter the optimal features. Airway-ANN and Basic-ANN models were established based on all collected indicators and remaining variables excluding airway assessment indicators, respectively. The performance of Basic-ANN, Airway-ANN and STOP-BANG was evaluated by the area under the precision-recall curve (AUPRC) on temporal validation set. The SHAP was used for revealing the predictive behavior of our best model.

Results

999 patients were eventually included. The AUPRC value of Airway-ANN model was significantly higher than Basic-ANN model in the temporal validation set (0.532 vs 0.429, P < 0.05). And the performance of both two ANN models was significantly better than that of STOP-BANG score (both P < 0.05). The Airway-ANN model was deployed to the cloud (http://njfh-yxb.com.cn:2022/airway_ann).

Conclusion

Our online interpretable Airway-ANN model achieved satisfying ability in identifying the hypoxemia risk in adult (18-64) patients undergoing EGD.

Incidence of oxygen desaturation using a high-flow nasal cannula versus a facemask during flexible bronchoscopy in patients at risk of hypoxemia: a randomised controlled trial

BACKGROUND

Patients with obstructive sleep apnoea (OSA), male sex, obesity, older age or hypertension are prone to hypoxemia during flexible bronchoscopy. This study investigated whether using a high-flow nasal cannula (HFNC) could reduce the incidence of oxygen desaturation during bronchoscopy under deep sedation in patients at risk of hypoxemia.

METHODS

A total of 176 patients at risk of hypoxemia who underwent flexible bronchoscopy under deep sedation were randomly assigned to two groups: the HFNC group (humidified oxygen was supplied via a high-flow nasal cannula at a rate of 60 L/min and a concentration of 100%, n = 87) and the facemask group (oxygen was supplied via a tight-fitting facemask at a rate of 6 L/min and a concentration of 100%, n = 89).

RESULTS

Oxygen desaturation occurred in 4 (4.6%) patients in the HFNC group and 26 (29.2%) patients in the facemask group (P < 0.001). The facemask group required more jaw thrust manoeuvres than the HFNC group (43[48.3%] vs. 5[5.7%], P < 0.001). 8 patients (9.0%) in the facemask group and none in the HFNC group required bag-mask ventilation (P = 0.012).

CONCLUSION

The use of an HFNC can reduce the incidence of oxygen desaturation and the requirement for airway intervention in patients at risk of hypoxemia during flexible bronchoscopy under deep sedation.

TRIAL REGISTRATION

www.chiCTR.org.cn Identifier: ChiCTR2100044105. Registered 11/03/2021.

High-Flow vs. Low-Flow Nasal Cannula in Reducing Hypoxemic Events During Bronchoscopic Procedures: A Systematic Review and Meta-Analysis

Introduction

High-flow nasal cannula (HFNC) oxygenation method has been proven to be successful in oxygenation of patients with respiratory failure and has exhibited clinical superiority compared to low-flow nasal cannula (LFNC).

Methods

We performed a systematic review and meta-analysis to evaluate the potential favorable impact of HFNC oxygenation during bronchoscopy and related procedures like endobronchial ultrasound-transbronchial needle aspiration. Only randomized control trials (RCTs) were included in the meta-analysis.

Results

Six randomized control trials with 1,170 patients were included in this meta-analysis. Patients who underwent bronchoscopy with the use of high-flow nasal cannula experienced less hypoxemic events/desaturations, less procedural interruptions and pneumothoraxes compared to patients under low-flow nasal cannula treatment. This beneficial effect of HFNC in hypoxemic events was persistent 10 min after the end of procedure.

Conclusion

The high-flow nasal cannula (HFNC) oxygenation method could reduce hypoxemic events and related peri- and post-bronchoscopic complications.

The effectiveness of high-flow nasal cannula during sedated digestive endoscopy: a systematic review and meta-analysis

BACKGROUND

Studies evaluating the role of high-flow nasal cannula (HFNC) in sedated digestive endoscopy have reported conflicting results. This meta-analysis evaluates the effectiveness of HFNC in patients undergoing digestive endoscopy procedures under sedation.

METHODS

PubMed, Medline, EMBASE, Cochrane Central Register of Controlled Trials and Web of Science, were searched from inception to 31 July 2021. Only randomized clinical trials comparing HFNC with standard nasal cannula oxygen (SNC) during sedated digestive endoscopy were included. The incidence of hypoxemia was the primary outcome, and the secondary outcome was the percentage of patients who needed airway interventions.

RESULTS

Seven studies that enrolled 2998 patients were included. When compared to SNC, HFNC was associated with a significant reduction in hypoxemia incidence (OR 0.24, 95% CI 0.09 to 0.64) and airway intervention requirements (OR 0.15, 95% CI 0.03 to 0.69), with both high heterogeneity (I² = 81% and 94%). In subgroup analysis, when hypoxemia was defined as pulse oxygen saturation (SpO₂) < 90%, low risk of hypoxemia subjects who received HFNC were associated with a significant reduction in hypoxemia incidence (OR 0.02, 95% CI 0.00 to 0.07; heterogeneity I² = 39%) and airway intervention requirements (OR 0.02, 95% CI 0.01 to 0.04; heterogeneity I² = 15%). However, in the high risk of hypoxemia subjects, there were no significant differences between the two oxygen administration techniques in both primary (OR 0.81, 95% CI 0.36 to 1.78; heterogeneity I² = 0%) and secondary outcomes (OR 0.85, 95% CI 0.46 to 1.59; heterogeneity I² = 0%).

CONCLUSIONS

Compared to SNC, HFNC not only reduce the incidence of hypoxemia but also reduce the requirements for airway interventions during sedated digestive endoscopy procedures, especially in patients at low risk for hypoxemia. In high risk of hypoxemia patients, there were no significant differences between the two oxygen administration techniques. Trial registration PROSPERO International prospective register of systematic reviews on 28 July 2021, registration no. CRD42021264556.

Oxygen desaturation during flexible bronchoscopy with propofol sedation is associated with sleep apnea: the PROSA-Study

BACKGROUND

Obstructive sleep apnea (OSA) is characterized by repetitive episodes of complete or partial obstruction of the upper airways during sleep. Conscious sedation for flexible bronchoscopy (FB) places patients in a sleep-like condition. We hypothesize that oxygen desaturation during flexible bronchoscopy may help to detect undiagnosed sleep apnea.

METHODS

Single-centre, investigator-initiated and driven study including consecutive patients undergoing FB for clinical indication. Patients completed the Epworth Sleepiness Scale (ESS), Lausanne NoSAS score, STOP-BANG questionnaire and the Berlin questionnaire and underwent polygraphy within 7 days of FB. FB was performed under conscious sedation with propofol. Oxygen desaturation during bronchoscopy was measured with continuous monitoring of peripheral oxygen saturation with ixTrend (ixellence GmbH, Germany).

RESULTS

145 patients were included in the study, 62% were male, and the average age was 65.8 ± 1.1 years. The vast majority of patients (n = 131, 90%) proved to fulfill OSA criteria based on polygraphy results: 52/131 patients (40%) had mild sleep apnea, 49/131 patients (37%) moderate sleep apnea and 30/131 patients (23%) severe sleep apnea. Patients with no oxygen desaturation had a significantly lower apnea-hypopnea index than patients with oxygen desaturation during bronchoscopy (AHI 11.94/h vs 21.02/h, p = 0.011). This association remained significant when adjusting for the duration of bronchoscopy and propofol dose (p = 0.023; 95% CI 1.382; 18.243) but did not hold when also adjusting for age and BMI.

CONCLUSION

The severity of sleep apnea was associated to oxygen desaturation during flexible bronchoscopy under conscious sedation. Patients with oxygen desaturation during bronchoscopy might be considered for sleep apnea screening.

TRIAL REGISTRATION

The Study was approved by the Ethics Committee northwest/central Switzerland, EKNZ (EK 16/13) and was carried out according to the Declaration of Helsinki and Good Clinical Practice guidelines. Due to its observational character, the study did not require registration at a clinical trial registry.